Alterations in Micro-Ribonucleic Acid Expression ... - Semantic Scholar

0013-7227/08/$15.00/0 Printed in U.S.A.

Endocrinology 149(4):1687–1696 Copyright © 2008 by The Endocrine Society doi: 10.1210/en.2007-0969

Alterations in Micro-Ribonucleic Acid Expression Profiles Reveal a Novel Pathway for Estrogen Regulation Amit Cohen, Michael Shmoish, Liraz Levi, Uta Cheruti, Berta Levavi-Sivan, and Esther Lubzens Department of Marine Biology (A.C., L.L., U.C., E.L.), Israel Oceanographic and Limnological Research, Haifa 31080, Israel; Department of Animal Sciences (A.C., L.L., B.L.-S.), Faculty of Agricultural, Food, and Environmental Quality Sciences, The Hebrew University of Jerusalem, Rehovot 76100, Israel; and Bioinformatics Knowledge Unit (M.S.), The Lorry I. Lokey Interdisciplinary Center for Life Sciences and Engineering, Technion-IIT, Haifa 32000, Israel Estrogens are steroid hormones that have been implicated in a variety of cellular and physiological processes in the development of diseases such as cancer and are also known to be associated with the effects of endocrine disrupting chemicals. Here we show that 17␤-estradiol (E2) alters microRNA (miRNA) expression profiles in the adult zebrafish (Danio rerio). An association between E2 and the expression of 25 miRNAs was found 12 h after treatment. Among the most up-regulated miRNAs were miR-196b and let-7h, and the most downregulated miRNAs included miR-130c and miR-101a. Tissuespecific changes in the transcripts levels of estrogen receptors (Esr1, Esr2a, and Esr2b) and miRNAs were found after hor-

E

STROGENS ARE STEROID hormones that have wide effects on different cellular and physiological processes (1, 2) and are known to be involved in the development of diseases such as cancer (3). Despite the existence of a significant body of work on gene regulation by 17␤-estradiol (E2), no comprehensive research was conducted on the regulation of microRNAs (miRNAs) by estrogens. miRNAs are 18- to 25-nt noncoding segments of RNA that negatively regulate gene expression at the posttranscriptional level and act in the fine-tuning of gene functions (4). Little is known about signals that induce miRNAs expression and there are only few examples for hormonal regulation of miRNAs (5), but some of them are associated with human disease, including cancer. Variability in miRNA expression profiles is associated with different estrogen receptor (ER) phenotypes of breast cancer (6, 7), raising the question whether ERs and their ligand E2 regulate the expression of miRNAs. This question is of importance as specific attention has been drawn to the effect of endocrine-disrupting chemicals, acting through estrogenic pathways (8). In oviparous vertebrates, including fish, the formation of eggs, a process known as vitellogenesis, consists of highly regulated pathways involving E2 (9). E2 regulates the synthesis of vitellogenins (Vtgs) in the liver of females but also the liver of E2-treated males First Published Online December 20, 2007 Abbreviations: E2, 17␤-Estradiol; ER, estrogen receptor; ERE, estrogen response element; miRNA, microRNA; nt, nucleotides; UTR, untranslated region; Vtg, vitellogenin; ZF, zebrafish. Endocrinology is published monthly by The Endocrine Society (http:// www.endo-society.org), the foremost professional society serving the endocrine community.

mone treatment. The most up-regulated miR-196b and its precursors are highly expressed in the skin and showed similar tissue-specific expression patterns after treatment, indicating a common pattern of regulation by E2. MiR-196b was shown to fine-tune the expression of its target gene Hoxb8a after treatment in whole-body homogenates. Taken together, our results suggest a novel pathway for the multifunctional and pleiotropic effects of estrogens and open new directions for future investigations of their association with miRNAs involved in estrogen-regulated physiological processes and diseases. (Endocrinology 149: 1687–1696, 2008)

responds by synthesizing Vtgs and other proteins associated with vitellogenesis (10). Most data to date supports the premise that estrogen action is mediated principally through the classical mechanism, in which estrogens diffuse into the cell and bind to ERs. Subsequently binding occurs in specific palindromic sequences of estrogen response elements (EREs), in the promoter region of estrogen-responsive genes (11, 12). To date, two ERs (ER␣ and ER␤1, designated as Esr1 and Esr2b, respectively, in zebrafish) have been characterized in mammals, birds, amphibians, and fish, and a third receptor, ER␤2 (designated as Esr2a in zebrafish), was discovered in teleost fish (13). Clear distinctions were found in the expression and tissue distribution of the different ERs, suggesting different functions for each ER subtype (14, 15). Here we show the association between E2 and miRNAs in vivo in the zebrafish (ZF), in which about 250 mature miRNAs have been identified; many of them were found to be evolutionarily conserved in other animals and show tissue specific expression patterns (16, 17). Materials and Methods E2 treatment and sample collection Adult ZF (3 months old) were purchased from a commercial ZF culture facility (A & H Holdings Ltd., Beit Itzhak, Israel) and maintained under conditions of a 14-h light, 10-h dark cycle. All fish were anesthetized with Tricaine (Sigma, St. Louis, MO) before experimental procedures (18), adhering to institutional regulations. ZF males were exposed to E2 (Sigma) by immersion (19). The concentration used was 5 ␮g/liter (18 nm) because 3– 4 ng/ml E2 was determined as the natural concentration in the plasma of adult vitellogenic female ZF (20). Ten fish were kept in each aquarium (6 liters in volume). Fish and tissue samples were collected after treatment at each time point (0, 1, 4, 12, and 24 h). The levels of E2 in the plasma were

1687 Downloaded from endo.endojournals.org at Hebrew University, Medical Library, 31253 on March 24, 2008

1688

Endocrinology, April 2008, 149(4):1687–1696

determined in a separate experiment in which samples (at 0, 4, 12, and 24 h) were collected by heparinized capillary tubes (VWR, West Chester, PA) from males exposed to 5 ␮g/liter E2 by immersion. E2 concentration in the plasma was determined by ELISA (estradiol enzyme immunoassay kit; Cayman Chemicals, Ann Arbor, MI), according to the manufacturer’s instructions. The sex of each sampled fish was confirmed by dissection. Samples were frozen instantly in liquid nitrogen and stored at ⫺80 C.

Cohen et al. • Estrogen Alters miRNA Expression Profiles

10-bp OrangeRuler marker (Fermentas, Hanover, MD) were analyzed on ethidium bromide-stained 12% polyacrylamide gel.

miRNA microarray miRNA microarray analysis was carried out by LC Sciences (Houston, TX; http://www.lcsciences.com/). Hybridization experiments were performed on RNA that was extracted from ZF whole-body homogenates. In brief, 5 ␮g of total RNA from untreated and E2-treated ZF

RNA extraction, polyadenylation, and reverse transcription RNA extraction was carried out using Trizol reagent (Invitrogen, Carlsbad, CA). Fish were weighed and 1 ml Trizol per 100 mg of fish weight was added, and the samples were homogenized. Four tissues were collected from four different individuals, and 1 ml Trizol was added to one tube containing the pooled samples. Polyadenylation and reverse transcription of RNA were carried out as described before (21). In brief, total RNA (5 ␮g) was treated with RQ1 deoxyribonuclease I (Promega, Madison, WI), and after phenol-chloroform extraction and ethanol precipitation, the treated total RNA was polyadenylated by Poly(A) polymerase at 37 C for 1 h, using the A-plus Poly(A) polymerase tailing kit (Epicentre, Madison, WI). The RNA was reverse transcribed using 200 U SuperScript II reverse transcriptase (Invitrogen) and 1 ␮g poly(T) adapter (GCGAGCACAGAATTAATACGACTCACTATAGGTTTTTTTTTTTTVN) at 42 C for 1 h.

Semiquantitative RT-PCR Total RNA was reverse transcribed using oligo(dT) primer (dT 23 VN) and Bio-RT reverse transcriptase (Bio-Lab Ltd., Jerusalem, Israel). Amplification of mRNAs was carried out with specific sets of primers for Esr1, Vtg3, and Ef1␣ (supplemental Table S1, published as supplemental data on The Endocrine Society’s Journals Online Web site at http:// endo.endojournals.org), using Ready Mix (Bio-Lab) for 35 cycles with a final annealing temperature of 60 C. The PCR products were analyzed on ethidium bromide-stained 1.5% agarose gel.

Primer design and real-time RT-PCR Total RNA was isolated, polyadenylated, and reverse transcribed as described before. The PCR mixture consisted of 1 ␮l of cDNA sample, 70 nm of each primer, and 12.5 ␮l of SYBR Green mix (Abgene, Epsom, UK), in a final volume of 25 ␮l. Amplification was carried out in a GeneAmp 5700 thermocycler (PE Applied Biosystems, Foster City, CA) under the following conditions: initial denaturation at 95 C for 15 min, followed by 40 cycles of denaturation at 95 C for 15 sec and annealingextension at 60 C for 1 min. Amplification of cDNAs was preformed in triplicates and sd of mean values was lower than 0.16 in all determinations. The relative abundance of ERs and Hox mRNAs were normalized to the amount of the elongation factor (Ef1␣) (22) using the equation: 2 ⫺ ⌬CT with ⌬CT ⫽ (CT ER ⫺ CT EF). ⌬CT corresponds to the difference between the CT (cycle threshold) measured for the target gene and the CT measured for the reference gene. Expression of miRNAs was calculated using the formula 2 ⫺ ⌬CT with ⌬CT ⫽ (CT miRNA ⫺ CT reference RNA), whereas in addition to Ef1␣, U11, and 7SK, small nuclear RNAs also served as reference genes because more than one reference gene provides additional accuracy to the calculated normalizations (23). Analysis of real-time RT-PCR data was performed using the REST-384 ␤ version 2 software (24). For the detection of mature miRNAs, the forward-specific primer sequence was the same as the mature miRNA sequence (21). Sequences were derived from the Sanger Institute miRBase (release 8.1) and Kloosterman et al. (17). The reverse primer (GCGAGCACAGAATTAATACGAC) in the amplification was fixed in all the reactions (21). For amplification of miR-196b precursors, primers were designed to the miRNA hairpin structure, as in Schmittgen et al. (25). Because the hairpin is contained within both the pri-miRNA and the pre-miRNA, primers designed to the hairpin should simultaneously amplify both RNAs (25). The list of primers is shown in supplemental Table S1. All primers were designed by Primer3 program (http:// frodo.wi.mit.edu/cgi-bin/primer3/primer3_www.cgi) and purchased from Sigma. They were validated for amplification efficiency and resulted in a single band by gel electrophoresis. The PCR products and the

FIG. 1. Induced expression of ERs and Vtg after E2 treatment. A, E2 concentrations in the plasma of control and E2-treated male fish (mean ⫾ SEM; n ⫽ 4). B, Results shown are of semiquantitative RTPCR, using primers sets for Esr1, Vtg3, and Ef1␣ (elongation factor 1␣) at different time points after E2 treatment. RNA was extracted from ZF whole-body homogenates, and PCR products were analyzed on ethidium bromide-stained 1.5% agarose gel. C, Relative real-time RT-PCR results of Esr1, Esr2a, and Esr2b mRNA levels (mean ⫾ SEM; n ⫽ 3) at different time points after E2 treatment. The expression was calculated relative to Ef1␣ as a reference gene. RNA was extracted from whole-body homogenates of control and treated fish. Asterisks represent statistically significant differences (one-way ANOVA, StudentNewman-Keuls post hoc test; **, P ⬍ 0.01).

Downloaded from endo.endojournals.org at Hebrew University, Medical Library, 31253 on March 24, 2008


Endocrinology, April 2008, 149(4):1687–1696 1689

FIG. 2. Alterations in miRNA expression profiles after E2 treatment. A, Heat map representation of differentially regulated miRNAs at different time points after E2 treatment (log2 scale; t test, P ⬍ 0.01). Red denotes high expression and green low expression. The range of expression values is from ⫺3-fold to ⫹3-fold; however, the expression map is saturated, and thus, the true variation in expression may be greater. Hybridization experiments were performed by LC Sciences on RNA that was extracted from ZF whole-body homogenates. Microarray results were hierarchically clustered using the GeneCluster program. Microarray data were submitted to ArrayExpress (accession no. E-MEXP-1147). B, Microarray analysis results of 38 ZF miRNAs that were differentially expressed (one-way ANOVA, Student-Newman-Keuls post hoc test, P ⬍ 0.05). The values on the y-axis are normalized expression ratios (fold scale) of the mean of three biological replicates. The values are presented in supplemental Table 2. C, Validation of microarray results by real-time RT-PCR. The expression profiles of eight selected miRNAs 12 h after treatment are presented in log2 scale. Pearson correlation analysis shows a significant correlation of 69%.


1690



FIG. 3. Expression patterns of selected miRNAs and ERs in different tissues. A, Relative real-time RT-PCR results of selected miRNAs in four tissues of untreated fish (pooled samples from four fish). Expression levels of selected tissues are represented relative to the tissue that shows the lowest expression level. B, Relative real-time RTPCR results of selected miRNAs in four tissues (pooled samples from four fish) 24 h after E2 treatment. C, Relative real-time RT-PCR results of Esr1, Esr2a, and Esr2b mRNA levels in four tissues (pooled samples from four fish) 24 h after E2 treatment. Also shown are the results of wholebody homogenates 24 h after E2 treatment, as shown in Fig. 1C. dre-miR, Danio rerio microRNA.




were size fractionated [⬍300 nucleotides (nt)] by YM-100 Microcon centrifugal filter (Millipore, Bedford, MA), and Poly(A) tails were added to the RNA sequences at the 3⬘ ends using a Poly(A) polymerase, and nucleotide tags were then ligated to the Poly(A) tails for later fluorescent dye staining. Dual-sample array assays were used, in which two sets of RNA sequences were added with tags of two different sequences and then combined together. The RNA sequences were hybridized overnight in a ␮Paraflo microfluidic miRNA microarray chip using a microcirculation pump. Tag-specific Cy3 and Cy5 staining dyes were then circulated through the microfluidic chip to complete labeling. Fluorescence images were collected using a laser scanner. In this study, each miRNA microarray chip contained 268 ZF mature miRNA probes [including miRNA* (miRNA star)] and 246 miRNA s-probes (most of these probes are directed to a partial miRNA* segment). Each miRNA probe had seven repeats. Array sequence contents were derived from the Sanger Institute miRBase (release 8.1) and Kloosterman et al. (17). Sequences and chip content were submitted to ArrayExpress (accession no. E-MEXP1147). For assay experimental design, RNA from one of three different treated fish (representing biological replicates) of each time point was hybridized with RNA from one of three different control fish in one miRNA chip. Dye swap was used to eliminate labeling-related biases.

Statistical data analysis Microarray data were analyzed by first subtracting the background and then normalizing the signals using a LOWESS filter (locally weighted regression). Normalized data were further analyzed by ANOVA using the software GeneSpring GX (version 7.3.1; Agilent Technologies, Santa Clara, CA). A miRNA to be listed as detectable must meet at least two conditions: signal intensity higher than three times (background sd) and spot coefficient of variation ⬍ 0.5, where coefficient of variation is calculated as (sd)/(signal intensity). In addition, where signals were detected for less than three of the repeats, they were considered unreliable and excluded from sets of detected miRNAs. In the dual-sample experiments, the ratio of the two sets of detected signals (log2 transformed, balanced) and P values of the t test were calculated. Differentially detected signals were those with P ⬍ 0.01. Microarray data were hierarchically clustered using the GeneCluster program (26). Dendrograms and expression maps were generated by the Treeview version 1.6 program (26). Statistical analysis was performed using the software XLSTAT (Addinsoft Inc., Paris, France).

Bioinformatics data analysis All miRNA sequences were obtained from miRbase database (27) (http://microrna.sanger.ac.uk/sequences/). A list of validated miRNA target genes was derived from the TarBase database (28) (http://www.diana.pcbi.upenn.edu/tarbase.html). The list of ZF ortholog genes and gene ontology was obtained from the Zebrafish Information Network database (29) (http://zfin.org/). Zebrafish 3⬘ untranslated region (UTR) sequences were obtained from the University of California, Santa Cruz, genome browser (30) (http://genome.ucsc.edu/). miRanda (version 1.0) software (31) was used for finding miRNA target sites in the 3⬘UTR sequences. Dragon ERE finder (version 2.0) program (32) was used for finding putative EREs in the ZF genome.

Results Revealing spatiotemporal expression patterns of ERs and miRNAs after E2 treatment

To elucidate whether E2 affects the expression profiles of miRNAs, we performed a time-course experiment at 0, 1, 4, 12, and 24 h, with whole-body homogenates and tissues obtained from E2-treated male ZF. The mean values for E2 in the plasma of fish ranged from 0.97 (controls) to 5.12 ng/ml in treated fish (Fig. 1A). A peak in the expression of Esr1 was observed in ZF whole-body homogenates 12 h after treatment (Fig. 1, B and C), coinciding with the expression of Vtg3 (Fig. 1B) and confirming the efficacy of the treatment. Microarray analysis (Fig. 2A), followed by ANOVA, revealed 38 differentially expressed miRNAs during different time points of the experiment, and 25 of these miRNAs were regulated mainly 12 h after treatment (Fig. 2B and supplemental Table S2). Among the most up-regulated miRNAs were miR-196b, let-7h, and let-7d and the most down-regulated miRNAs included miR-130c, miR-130a, and miR-101a (Fig. 2B and supplemental Table S2). Microarray results were verified by testing the expression of selected eight miRNAs in real-time RT-PCR (Fig. 2C). miRNA expression in the ZF is tissue dependent (17, 33). Therefore, the miRNA expression patterns were evaluated in four different tissues: the liver, intestine, skin, and gills (Fig. 3, A and B) because these tissues showed variable ER expression patterns in response to E2 treatment (Fig. 3C). Results show that the expression of Esr1 is most up-regulated in the liver (Fig. 3C), in which miR-19a, miR-101a, and miR122 are mostly expressed (Fig. 3A and Table 1). E2 treatment reduced the expression level of most miRNAs in the liver (Fig. 3B). Esr2a shows up-regulation in the skin, intestine, and gills (Fig. 3C). Esr2b shows relatively low up-regulation in the intestine and down-regulation in the other tissues (Fig. 3C). Whereas in the skin most miRNAs were up-regulated, the E2 treatment resulted in down-regulation of miRNA expression in the gills and almost no change (except for miR-196b) in the intestine. To evaluate the relative expression level of miRNAs in the different tissues (in untreated fish, Fig. 3A), the expression of each miRNA was set as 1 for the tissue showing the lowest expression level and the level in the other tissues was compared with the level in this tissue. For example, for miR-122, the expression level in the skin was set as 1 and progressively higher expression levels are shown in intestine, gills, and liver. We next compared the expression patterns of miR-196b to

TABLE 1. An overview of miRNA expression patterns in different tissues and their regulation by E2

a

miRNA

Expressed highest in tissue(s)

Expression pattern in cited publications (reference no.)

Regulation after E2

Regulation in other tissues

19a 101a 130c 460 –5p 122 let-7 h 29b*a 196b

Liver Liver Gills, skin Skin, gills Liver Gills, skin Gills Skin, intestine

Ubiquitous (33) Ubiquitous (33) Ubiquitous (17) Fins (17) Liver, gut (33) NE NE NE

Down NC Down Up Down Up Down Up/down

Down Up/down Down Up Up/down NC Up/down Up/down

, 29b* (partially complementary miRNA star). NE, Not examined; NC, not changed.


1692


the expression patterns of its precursors in untreated and treated fish. MiR-196b and its precursors are highly expressed in the skin and intestine tissues in untreated fish (Fig. 4A) and show similar tissue-specific expression patterns after E2 treatment (Fig. 4B). The expression patterns after E2 treatment showed up-regulation of miR-196b in the skin and down-regulation in the intestine and liver.


The relation between miR-196b and Hoxb cluster genes

Our results show that miR-196b is the most up-regulated miRNA in whole-body homogenates after E2 treatment (Fig. 2B and supplemental Table S2). Like other miR-196 family members (34), miR-196b is embedded in the Hoxb cluster in the ZF genome (Fig. 5A). To reveal whether there is an association among E2, miR-196b, and Hox genes in the adult ZF, Dragon ERE Finder program was used (32) and disclosed four putative EREs in the Hoxb cluster (Fig. 5A): 1) 8109 nt upstream of Hoxb10a (reverse complement strand: AGGATCC-TTA-TGACC-TG), 2) 7674 nt upstream of Hoxb8a (forward strand: CA-GCGGA-ATT-TGACC-AT), 3) 6120 nt downstream of Hoxb7a (reverse complement strand: CTCGTCC-ACA-TGACC-AT), and 4) 8795 nt downstream of Hoxb7a (forward strand: AA-AGTCG-TGT-TGACC-TT). A number of variations were found on the ERE consensus sequence of 5⬘-C(A/G)GGTCAnnnTGACC(T/C)G-3⬘ that accommodate stable ER binding (35). The expression patterns of four hox genes after E2 treatment show that Hoxb8a is up-regulated 12 h after treatment in whole-body homogenates, using real-time RT-PCR (Fig. 5B). Hoxb8a induction coincides with the up-regulation of Hoxb7a, another member of this cluster, but there is no significant induction of Hoxb9a or of Hoxb10a (Fig. 5B). The association of miR-196 and its target gene Hoxb8a after E2 treatment

It is known that miR-196 regulates Hoxb8 by a directed cleavage mechanism (36). To further investigate the mode of regulation of miR-196b, we designed two primer sets (Fig. 6A), in which one primer set encompasses the miR-196 binding and cleavage sites, and the other primer set is located 35 nt upstream of the miR-196 binding site and will therefore amplify the polyadenylated 5⬘ segment of the cleaved Hoxb8a mRNA. Results show a significant decrease in Hoxb8a expression in whole-body homogenates after E2 treatment with the primer set that encompasses miR-196b cleavage site (Fig. 6B), indicating that miR-196b repressed Hoxb8a expression. However, E2 treatment did not change the Hoxb8a expression in the skin, possibly due to the general low level of expression of Hoxb8a in this tissue (data not shown). Discussion

FIG. 4. Expression patterns of miR-196b and its precursors in different tissues. A, Relative real-time RT-PCR results of miR-196b (mature form and precursors) in four different tissues of untreated fish (pooled samples from four fish). Expression levels of selected tissues are represented relative to the liver. B, Relative real-time RT-PCR results of miR-196b (mature form and precursors) in four different tissues (pooled samples from four fish) 24 h after E2 treatment. C, Amplicons of real-time RT-PCR of miR-196b primers used during this study. PCR products were analyzed on ethidium bromidestained 12% polyacrylamide gel. The 66-bp product is the amplification result of 22 bp of a miR-196b specific primer together with 44 bp of the poly(T) adapter (21). The 58-bp products are the amplification result of miR-196b precursors hairpin structures (25) (supplemental Table 1). dre-miR, Danio rerio microRNA.

In vertebrates, E2 regulates reproduction but also a large array of physiological processes such as growth; differentiation; metabolism; homeostasis of male and female reproductive organs; and functioning of the skeletal, cardiovascular, immune, and central nervous systems (1–3). Our results show that E2 regulates the expression of miRNAs and thus contributes to improve our understanding of the wide pleiotropic effects of estrogens. E2 affects the expression of several miRNAs in the adult ZF, probably through Esr1 and/or Esr2a. Esr1 and Esr2a are the only ER genes that were significantly induced together with Vtg3 after E2 treatment (Fig. 1, B and C). These ERs are also known to be responsible for the regulation of other Vtg genes (37). A variable pattern of ERs expression was found in the various tissues, with Esr1 mostly up-regulated in the liver and Esr2a in the skin, in-




FIG. 5. Genome location of miR-196b and analysis of Hoxb cluster genes after E2 treatment. A, A detailed overview of the ZF Hoxb cluster region as appears in Ensembl Zebrafish (www.ensembl.org/index.html). Arrowheads beside gene names show the direction of transcription. A colored box represents each exon, whereas angled lines represent introns. The location of the miR-196b gene is represented by a dashed line. Asterisks show the location of putative EREs as were predicted by the Dragon ERE Finder program (32) (http://sdmc.lit.org.sg/ERE-V2/index). B, The mRNA levels of Hoxb7a, Hoxb8a, Hoxb9a, and Hoxb10a (mean ⫾ SEM; n ⫽ 3) were measured by relative real-time RT-PCR in ZF whole-body homogenates 12 h after E2 treatment. The differences in the mean values among Hoxb7a and Hoxb8a vs. Hoxb9a were significant (one-tailed t test; *, P ⬍ 0.05). dre-miR, Danio rerio microRNA.

testine, and gills, supporting previously published results in fish (38). The regulation of Esr2b is significantly lower than that of the other ERs (Figs. 1B and 3C) (38). The pattern of miRNAs expression changed after E2 treatment (Fig. 3B), with the down-regulation of most of the miRNAs in the liver and their up-regulation in the skin. These results may reflect events occurring at the onset or during vitellogenesis because E2 levels in the plasma of treated males (Fig. 1A) were within the range found in vitellogenic females (20). Regulation of miRNAs is expected to be associated with functional changes of their target genes. For example, it has

been reported that the liver-specific miR-122 modulates lipid metabolism through regulation of genes that are involved in this process (39). Additional putative target genes could be identified through the discovery of evolutionary conserved miRNA binding sites in their 3⬘UTRs (supplemental Table S3). Whereas only a relatively small number of miRNA target genes have been verified so far in animals, computational predictions show that each miRNA can potentially regulate the expression of hundreds of different genes (27). It can therefore be assumed that the observed alterations in miRNA expression profiles after E2 treatment lead to changes in a


1694



FIG. 6. Hoxb8a mRNA is cleaved at miR-196 binding site after E2 treatment. A, Location of the primer sets on Hoxb8a mRNA 3⬘UTR sequence. The first primer set encompassing miR-196 binding site consists of the forward primer and reverse primer 2. The second primer set consists of the forward primer and reverse primer 1. B, Relative real-time RT-PCR analysis of Hoxb8a mRNA levels (mean ⫾ SEM; n ⫽ 3) in whole-fish homogenates 12 h after treatment relative to untreated fish, using the two different primer sets for Hoxb8a mRNA. The difference in the mean value between the two PCR products was significant (one-tailed t test; *, P ⫽ 0.02).

large array of miRNA target genes, thereby contributing to the pleiotropic effects of estrogens on different cellular and physiological processes. In most animals, miRNAs are believed to act by repressing translation (4), and recent studies have demonstrated that miRNAs also induce mRNA degradation or cleavage, even if the target sites have incomplete complementarity to the miRNA (40). Therefore, an approach that combines miRNA expression profiles together with transcriptome and proteome analyses of E2-treated fish will be required to elucidate the global role of miRNAs in estrogen regulation pathways (41). A detailed study of the expression of miR-196b, the most up-regulated miRNA after E2 treatment, revealed its close association with Hoxb8a, one of the validated target genes of miR-196b (supplemental Table S3). In the adult ZF, induction of miR-196b may serve for fine-tuning regulation of Hoxb8a expression after E2 treatment. This type of loop (Fig. 7) was described before regarding the role of miRNAs in the canalization of genetic networks (42). The induction of Hoxb8a after E2 treatment coincided with the up-regulation of Hoxb7a, another member of this cluster, but there was no induction of Hoxb9a (Fig. 5B). This pattern of regulation is reminiscent of observations made by Mansfield et al. (43), in which miR-196 was shown to have a Hox-like expression pattern during mouse embryonic development but differed from those of its nearest neighbors (Hox9-group genes). It is worth mentioning that during mouse embryonic development, miR-196 was found to inhibit Hoxb8 expression after treatment of retinoic acid (44), the ligand of the retinoic acid receptor, another member of the nuclear hormone receptor superfamily (45). In this study (44), a key morphogen, Sonic

Hedgehog, was the Hoxb8 downstream target, regulated by miR-196. Interestingly, miR-214, another E2 up-regulated miRNA (supplemental Table S2) was shown recently to modulate Hedgehog signaling during ZF embryonic development (46). Several hormones, including E2, were shown to regulate Hox genes expression and thereby mediate developmental processes in the embryo as well as functional differentiation in the adult mammalian organism (47). Because our results and the results of others (48) have shown only Hox gene expression in whole-body homogenates of adult ZF, their function and pattern of expression in specific tissues of the adult organism remain to be shown. Whereas our results indicate a mode of gene regulation by E2 in adult fish, they

FIG. 7. A model describing the relations between E2 treatment and the expression of miR-196b and Hoxb8a genes in the adult ZF.



also suggest an association among ERs, miRNAs, and Hox genes during different stages of animal development (14, 49, 50). The exact mechanisms for miRNAs regulation by estrogen still have to be elucidated and may include regulation at the transcriptional and pri/pre-miRNA processing levels. The similar expression patterns of the mature and the precursors of miR-196b suggest the occurrence of regulation at the transcriptional level (25). It will be interesting to reveal whether miR-196b is regulated by E2 also in human cells, especially because there is evidence that miR-196 genes are among the induced miRNA genes in human breast (6), lung and prostate (51), and pancreatic cancers (52). The observation that miR196b was the most up-regulated miRNA after E2 treatment could be related to the fact that it was highly expressed in the skin. Therefore, miR-196b may also be used as a biomarker of exposure to environmental estrogens and endocrine-disrupting chemicals that fish may encounter in their aquatic environment. Acknowledgments We thank Dr. J. Cerda` (Lab Institut de Recerca i Tecnologia Agroalimentaries-Institut de Cie`ncies del Mar Centre Mediterrani d’Investigacions Marines i Ambientals, Barcelona, Spain) and Professor P. W. H. Holland (Department of Zoology, University of Oxford, Oxford, UK) for critically reading of the manuscript. We also thank Irena Pekarsky and Aliza Hadani for technical assistance and Hana Bernard for drawing the figures. Received July 16, 2007. Accepted December 7, 2007. Address all correspondence and requests for reprints to: Dr. Esther Lubzens, Department of Marine Biology, Israel Oceanographic and Limnological Research, Haifa 31080, Israel. E-mail: [email protected]. This work was supported by Israel Science Foundation Grant 1184. Marine Genomics Europe (MGE; FP6-EU Network of Excellence Grant GOCE-CT-2004-505403) was performed within the framework of the Fish Cells (FICEL) flagship project of MGE. Disclosure Statement: The authors of this manuscript have nothing to declare.

References 1. Korach KS 1994 Insights from the study of animals lacking functional estrogen receptor. Science 266:1524 –1527 2. Moggs JG, Orphanides G 2001 Estrogen receptors: orchestrators of pleiotropic cellular responses. EMBO Rep 2:775–781 3. Deroo BJ, Korach KS 2006 Estrogen receptors and human disease. J Clin Invest 116:561–570 4. Bartel DP 2004 MicroRNAs: genomics, biogenesis, mechanism, and function. Cell 116:281–297 5. Cuellar TL, McManus MT 2005 MicroRNAs and endocrine biology. J Endocrinol 187:327–332 6. Iorio MV, Ferracin M, Liu CG, Veronese A, Spizzo R, Sabbioni S, Magri E, Pedriali M, Fabbri M, Campiglio M, Me´nard S, Palazzo JP, Rosenberg A, Musiani P, Volinia S, Nenci I, Calin GA, Querzoli P, Negrini M, Croce CM 2005 MicroRNA gene expression deregulation in human breast cancer. Cancer Res 65:7065–7070 7. Mattie MD, Benz CC, Bowers J, Sensinger K, Wong L, Scott GK, Fedele V, Ginzinger D, Getts R, Ha C 2006 Optimized high-throughput microRNA expression profiling provides novel biomarker assessment of clinical prostate and breast cancer biopsies. Mol Cancer 5:24 8. Tabb MM, Blumberg B 2006 New modes of action for endocrine-disrupting chemicals. Mol Endocrinol 20:475– 482 9. Arukwe A, Goksoyr A 2003 Eggshell and egg yolk proteins in fish: hepatic proteins for the next generation: oogenetic, population, and evolutionary implications of endocrine disruption. Comp Hepatol 2:2 10. Denslow ND, Lee HS, Bowman CJ, Hemmer MJ, Folmar LC 2001 Multiple responses in gene expression in fish treated with estrogen. Comp Biochem Physiol B Biochem Mol Biol 129:277–282 11. Hall JM, Couse JF, Korach KS 2001 The multifaceted mechanisms of estradiol and estrogen receptor signaling. J Biol Chem 276:36869 –36872


12. Gruber CJ, Gruber DM, Gruber IM, Wieser F, Huber JC 2004 Anatomy of the estrogen response element. Trends Endocrinol Metab 15:73–78 13. Hawkins MB, Thornton JW, Crews D, Skipper JK, Dotte A, Thomas P 2000 Identification of a third distinct estrogen receptor and reclassification of estrogen receptors in teleosts. Proc Natl Acad Sci USA 97:10751–10756 14. Bardet PL, Horard B, Robinson-Rechavi M, Laudet V, Vanacker JM 2002 Characterization of oestrogen receptors in zebrafish (Danio rerio). J Mol Endocrinol 28:153–163 15. Menuet A, Pellegrini E, Anglade I, Blaise O, Laudet V, Kah O, Pakdel F 2002 Molecular characterization of three estrogen receptor forms in zebrafish: binding characteristics, transactivation properties, and tissue distributions. Biol Reprod 66:1881–1892 16. Chen PY, Manninga H, Slanchev K, Chien M, Russo JJ, Ju J, Sheridan R, John B, Marks DS, Gaidatzis D, Sander C, Zavolan M, Tuschl T 2005 The developmental miRNA profiles of zebrafish as determined by small RNA cloning. Genes Dev 19:1288 –1293 17. Kloosterman WP, Steiner FA, Berezikov E, de Bruijn E, van de Belt J, Verheul M, Cuppen E, Plasterk RH 2006 Cloning and expression of new microRNAs from zebrafish. Nucleic Acids Res 34:2558 –2569 18. Westerfield M 1995 The Zebrafish book. a guide for the laboratory use of zebrafish (Brachydanio rerio). 3rd ed. Eugene, OR: University of Oregon Press 19. Tong Y, Shan T, Poh YK, Yan T, Wang H, Lam SH, Gong Z 2004 Molecular cloning of zebrafish and medaka vitellogenin genes and comparison of their expression in response to 17␤-estradiol. Gene 328:25–36 20. Heiden TK, Carvan MJ, Hutz RJ 2006 Inhibition of follicular development, vitellogenesis, and serum 17␤-estradiol concentrations in zebrafish following chronic, sublethal dietary exposure to 2,3,7,8-tetrachlorodibenzo-p-dioxin. Toxicol Sci 90:490 – 499 21. Shi R, Chiang VL 2005 Facile means for quantifying microRNA expression by real-time PCR. Biotechniques 39:519 –525 22. Olsvik PA, Lie KK, Jordal AE, Nilsen TO, Hordvik I 2005 Evaluation of potential reference genes in real-time RT-PCR studies of Atlantic salmon. BMC Mol Biol 6:21 23. Vandesompele J, De Preter K, Pattyn F, Poppe B, Van Roy N, De Paepe A, Speleman F 2002 Accurate normalization of real-time quantitative RT-PCR data by geometric averaging of multiple internal control genes. Genome Biol 3:research0034 24. Pfaffl MW, Horgan GW, Dempfle L 2002 Relative expression software tool (REST) for group-wise comparison and statistical analysis of relative expression results in real-time PCR. Nucleic Acids Res 30:e36 25. Schmittgen TD, Jiang J, Liu Q, Yang L 2004 A high-throughput method to monitor the expression of microRNA precursors. Nucleic Acids Res 32:e43 26. Eisen MB, Spellman PT, Brown PO, Botstein D 1998 Cluster analysis and display of genome-wide expression patterns. Proc Natl Acad Sci USA 95: 14863–14868 27. Griffiths-Jones S, Grocock RJ, van Dongen S, Bateman A Enright AJ 2006 miRBase: microRNA sequences, targets and gene nomenclature. Nucleic Acids Res 34:D140 –D144 28. Sethupathy P, Corda B, Hatzigeorgiou AG 2006 TarBase: a comprehensive database of experimentally supported animal microRNA targets. RNA 12: 192–197 29. Sprague J, Bayraktaroglu L, Clements D, Conlin T, Fashena D, Frazer K, Haendel M, Howe DG, Mani P, Ramachandran S, Schaper K, Segerdell E, Song P, Sprunger B, Taylor S, Van Slyke CE, Westerfield M 2006 The Zebrafish Information Network: the zebrafish model organism database. Nucleic Acids Res 34:D581–D585 30. Karolchik D, Baertsch R, Diekhans M, Furey TS, Hinrichs A, Lu YT, Roskin KM, Schwartz M, Sugnet CW, Thomas DJ, Weber RJ, Haussler D, Kent WJ; University of California Santa Cruz 2003 The UCSC Genome Browser Database. Nucleic Acids Res 31:51–54 31. Enright AJ, John B, Gaul U, Tuschl T, Sander C, Marks DS 2003 MicroRNA targets in Drosophila. Genome Biol 5:R1 32. Bajic VB, Tan SL, Chong A, Tang S, Stro¨m A, Gustafsson JA, Lin CY, Liu ET 2003 Dragon ERE Finder version 2: a tool for accurate detection and analysis of estrogen response elements in vertebrate genomes. Nucleic Acids Res 31: 3605–3607 33. Wienholds E, Kloosterman WP, Miska E, Alvarez-Saavedra E, Berezikov E, de Bruijn E, Horvitz HR, Kauppinen S, Plasterk RH 2005 MicroRNA expression in zebrafish embryonic development. Science 309:310 –311 34. Tanzer A, Amemiya CT, Kim CB, Stadler, PF 2005 Evolution of microRNAs located within Hox gene clusters. J Exp Zoolog B Mol Dev Evol 304:75– 85 35. Driscoll MD, Sathya G, Muyan M, Klinge CM, Hilf R, Bambara RA 1998 Sequence requirements for estrogen receptor binding to estrogen response elements. J Biol Chem 273:29321–29330 36. Yekta S, Shih IH, Bartel DP 2004 MicroRNA-directed cleavage of HOXB8 mRNA. Science 304:594 –596 37. Menuet A, Le Page Y, Torres O, Kern L, Kah O, Pakdel F 2004 Analysis of the estrogen regulation of the zebrafish estrogen receptor (ER) reveals distinct effects of ER␣, ER␤1 and ER␤2. J Mol Endocrinol 32:975–986 38. Filby AL, Tyler CR 2005 Molecular characterization of estrogen receptors 1, 2a, and 2b and their tissue and ontogenic expression profiles in fathead minnow (Pimephales promelas). Biol Reprod 73:648 – 662


1696


39. Esau C, Davis S, Murray SF, Yu XX, Pandey SK, Pear M, Watts L, Booten SL, Graham M, McKay R, Subramaniam A, Propp S, Lollo BA, Freier S, Bennett CF, Bhanot S, Monia BP 2006 miR-122 regulation of lipid metabolism revealed by in vivo antisense targeting. Cell Metab 3:87–98 40. Lim LP, Lau NC, Garrett-Engele P, Grimson A, Schelter JM, Castle J, Bartel DP, Linsley PS, Johnson JM 2005 Microarray analysis shows that some microRNAs downregulate large numbers of target mRNAs. Nature 433:769 –773 41. Nakahara K, Kim K, Sciulli C, Dowd SR, Minden JS, Carthew RW 2005 Targets of microRNA regulation in the Drosophila oocyte proteome. Proc Natl Acad Sci USA 102:12023–12028 42. Hornstein E, Shomron N 2006 Canalization of development by microRNAs. Nat Genet 38:S20 –S24 43. Mansfield JH, Harfe BD, Nissen R, Obenauer J, Srineel J, Chaudhuri A, Farzan-Kashani R, Zuker M, Pasquinelli AE, Ruvkun G, Sharp PA, Tabin CJ, McManus MT 2004 MicroRNA-responsive ‘sensor’ transgenes uncover Hoxlike and other developmentally regulated patterns of vertebrate microRNA expression. Nat Genet 36:1079 –1083 44. Hornstein E, Mansfield JH, Yekta S, Hu JK, Harfe BD, McManus MT, Baskerville S, Bartel DP, Tabin CJ 2005 The microRNA miR-196 acts upstream of Hoxb8 and Shh in limb development. Nature 438:671– 674 45. Aranda A, Pascual A 2001 Nuclear hormone receptors and gene expression. Physiol Rev 81:1269 –1304 46. Flynt AS, Li N, Thatcher EJ, Solnica-Krezel L, Patton JG 2007 Zebrafish


47. 48.

49. 50.

51.

52.

miR-214 modulates Hedgehog signaling to specify muscle cell fate. Nat Genet 39:259 –263 Daftary GS, Taylor HS 2006 Endocrine regulation of HOX genes. Endocr Rev 27:331–355 Corredor-Ada´mez M, Welten MC, Spaink HP, Jeffery JE, Schoon RT, de Bakker MA, Bagowski CP, Meijer AH, Verbeek FJ, Richardson MK 2005 Genomic annotation and transcriptome analysis of the zebrafish (Danio rerio) hox complex with description of a novel member, hox b 13a. Evol Dev 7:362– 375 Pearson JC, Lemons D, McGinnis W 2006 Modulating Hox gene functions during animal body patterning. Nat Rev Genet 6:893–904 Tingaud-Sequeira A, Andre´ M, Forgue J, Barthe C, Babin PJ 2004 Expression patterns of three estrogen receptor genes during zebrafish (Danio rerio) development: evidence for high expression in neuromasts. Gene Expr Patterns 4:561–568 Volinia S, Calin GA, Liu CG, Ambs S, Cimmino A, Petrocca F, Visone R, Iorio M, Roldo C, Ferracin M, Prueitt RL, Yanaihara N, Lanza G, Scarpa A, Vecchione A, Negrini M, Harris CC, Croce CM 2006 A microRNA expression signature of human solid tumors defines cancer gene targets. Proc Natl Acad Sci USA 103:2257–2261 Bloomston M, Frankel WL, Petrocca F, Volinia S, Alder H, Hagan JP, Liu CG, Bhatt D, Taccioli C, Croce CM 2007 MicroRNA expression patterns to differentiate pancreatic adenocarcinoma from normal pancreas and chronic pancreatitis. JAMA 297:1901–1908

Endocrinology is published monthly by The Endocrine Society (http://www.endo-society.org), the foremost professional society serving the endocrine community.
